Papers by shubhangi sapkal
Human gestures pertaining to face are a usual way for humans to describe their feelings and emoti... more Human gestures pertaining to face are a usual way for humans to describe their feelings and emotions. This research aims to develop a facial recognition model and apply it to the current set of expressions. There are five senses: fear, pleasure, disgust, sadness and surprise using the innovative vector machine (SVM) and the local binary (LBP). In this work, we propose a new method of recognition of facial expression that uses a local binary pattern (LBP) and a local phase quantification (LPQ) based on Gabor's facial image. To capture the outstanding visual properties, the Gabor filter is first adopted to extract the characteristics of the face image between five scales and eight orientations. Then, the Gabor image is encoded by the LBP operator and the LPQ operator, respectively. The proposed algorithm is then compared with other algorithm viz: principal component analysis. The result shows that the proposed method outperforms many other approaches in this document in terms of accuracy and using Japanese Female Facial Expression Database.
Biometrie is a reliable and convenient way for person authentication. Template security is a majo... more Biometrie is a reliable and convenient way for person authentication. Template security is a major concern in practical implementation of biometric system. This work proposes cryptosystem for face verification with fuzzy vault. Here, a secrete value k is locked using fuzzy vault with a set of points. Fuzzy vault scheme is order invariant and applied for point set, hence SIFT is suitable feature extraction method. SIFT is distinctive invariant feature extraction method. As there is a trade-off between accuracy and security, this is a challenge to implement biometric template protection scheme which will give better accuracy of the biometric system.
Neural networks provide an obvious technique for classification. In this paper, neural network ap... more Neural networks provide an obvious technique for classification. In this paper, neural network approach is used for classification of Loan applications. Data has been collected from publicly available source, Alyuda Research, Inc.[1].It does not guarantee the accuracy of the data. This data is intended solely for experimental purposes. FFNN and RBN are used for classification. PREMNMX matlab Function Preprocesses data so that minimum is-1 and maximum is 1
This study focused on development and application of efficient algorithm for clustering and class... more This study focused on development and application of efficient algorithm for clustering and classification of supervised visual data. In machine learning clustering classification and clustering is most useful techniques in pattern recognition and computer vision. Existing techniques for subspace clustering makes use of rank minimization and spars based that are computationally expensive and may result in reducing the clustering performance. The algorithm called fixed rank representation based on matrix factorization is used to partially solve the problem of existing system. FRR perform classification using neural network algorithm. Neural network takes patterns as input from the FRR by performing rank minimization techniques in LRR and then performs classification. FRR can be able to solve the problem of multiple subspace clustering. Using neural network FRR can be able to give good classification of given pattern. Keywords— Supervised visual data classification, SSC, LRR, FFR, Neu...
Fingerprint techniques are widely used in authentication systems, therefore its privacy protectio... more Fingerprint techniques are widely used in authentication systems, therefore its privacy protection becomes an important issue. Securing a stored fingerprint template is very important because once fingerprints are compromised, it cannot be easily revoked. So, we review here a new system for preserving fingerprint confidentiality. In this system, the fingerprint privacy is maintained by combining two different fingerprints into a new identity. In the enrollment phase, two fingerprints are taken from two different fingers. We obtain the minutiae positions of one fingerprint, the orientation of another fingerprint, and the reference points from both fingerprints. Based on the obtained information, a combined minutiae template is generated and stored in a database. In the authentication phase, we use the fingerprints of the same fingers that are already used in enrollment phase. For matching the two query fingerprints against a combined minutiae template, a two-stage fingerprint matchin...
We present in this paper, Fourier descriptor and feedforward neural network for face recognition.... more We present in this paper, Fourier descriptor and feedforward neural network for face recognition. Analysis is done for various numbers of iterations. Comparison shows that faces are recognized with FFNN more accurately with 50000 iterations. For experiment, FERET database is used.
Automatic Physiognomy recognition systems are now popularly used in various applications ranging ... more Automatic Physiognomy recognition systems are now popularly used in various applications ranging from mobile payment verification to inbuild security access. The use of Physiognomy recognition has increased awareness about facial simulation attacks viz is also called as a biometric sensor presentation attack) that can use a picture or motion file of the Physiognomy of an known person with access to facilities or services. While the amount of Physiognomy identification methodologies that have been proposed do have the ability to place general implications however they are not adequately addressed that is why we offer a powerful and robust Physiognomy detection algorithm using image distortion analysis (IDA). Four different properties (Spectrum deflection, interval, color, and variety with colors) these can be separated to create an IDA class feature set vector, which consists of several SVM classifiers that have been trained for disguised Physiognomy forgery. (Eg printed photos and r...
2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2017
Biometrie is a reliable and convenient way for person authentication. Template security is a majo... more Biometrie is a reliable and convenient way for person authentication. Template security is a major concern in practical implementation of biometric system. This work proposes cryptosystem for face verification with fuzzy vault. Here, a secrete value k is locked using fuzzy vault with a set of points. Fuzzy vault scheme is order invariant and applied for point set, hence SIFT is suitable feature extraction method. SIFT is distinctive invariant feature extraction method. As there is a trade-off between accuracy and security, this is a challenge to implement biometric template protection scheme which will give better accuracy of the biometric system.
International Journal of Advanced engineering, Management and Science, 2017
As the cloud computing technology develops during the recent days, outsourcing data to cloud serv... more As the cloud computing technology develops during the recent days, outsourcing data to cloud service for storage becomes an attractive trend, which benefits in sparing efforts on heavy data maintenance and management. Nevertheless, since the outsourced cloud storage is not fully trustworthy, it raises security concerns on how to realize data deduplication in cloud while achieving integrity auditing. In this work, we study the problem of integrity auditing and secure deduplication on cloud data. Specifically, willing achieving both data integrity and deduplication in cloud, we propose two secure systems, namely SecCloud and SecCloud. SecCloud introduces an auditing entity with maintenance of a Map Reduce cloud, which helps clients generate data tags before uploading still audit the integrity of data having been stored in cloud. Compared with previous work, the computation by user in SecCloud is greatly reduced during the file uploading and auditing phases. SecCloud is designed motivated individually fact that customers always must encrypt their data before uploading, and enables integrity auditing and secure deduplication on encrypted data.
Zenodo (CERN European Organization for Nuclear Research), Sep 10, 2022
A Hipster recommender system proposed in this paper is a movie recommender system. The objective ... more A Hipster recommender system proposed in this paper is a movie recommender system. The objective of our system is to recommend those movies that are mostly unseen but have very good reviews thus the name-Hipster. Unlike other mentioned recommender systems that uses collaborative filtering or content based filtering or a combination of them and some other algorithms, this paper creates movie recommendations based on a hybrid algorithm that uses certain features of content based algorithm and a personalized popularity based algorithm. The paper also makes comparisons of proposed system and existing system based on attributes such as accuracy and variations of using different metrics. Thus we create a specialized movie recommender system that recommends only great movies of the genre that the user prefers and that which remains widely unknown, like hidden gems.
Proceedings of the Second International Conference on Information and Communication Technology for Competitive Strategies - ICTCS '16, 2016
Conventional security methods like password and ID card methods are now rapidly replacing by biom... more Conventional security methods like password and ID card methods are now rapidly replacing by biometrics for identification of a person. Biometrics uses physiological or behavioral characteristics of a person. Usage of biometric raises critical privacy and security concerns that, due to the noisy nature of biometrics, cannot be addressed using standard cryptographic methods. The loss of an enrollment biometric to an attacker is a security hazard because it may allow the attacker to get an unauthorized access to the system. Biometric template can be stolen and intruder can get access of biometric system using fake input. Hence, it becomes essential to design biometric system with secure template or if the biometric template in an application is compromised, the biometric signal itself is not lost forever and a new biometric template can be issued. One way is to combine the biometrics and cryptography or use transformed data instead of original biometric template. But traditional cryptography methods are not useful in biometrics because of intra-class variation. Biometric cryptosystem can apply fuzzy vault, fuzzy commitment, helper data and secure sketch, whereas, cancelable biometrics uses distorting transforms, Bio-Hashing, and Bio-Encoding techniques. In this paper, biometric cryptosystem is presented with fuzzy vault and fuzzy commitment techniques for fingerprint recognition system.
Many of the few biometric measures built up, the finger knuckle surface is turning into a famous ... more Many of the few biometric measures built up, the finger knuckle surface is turning into a famous selection of specialists because of its normal simplicity of reproducibility and confirmation. For any reason for individual distinguishing proof or wrong doing examination, finger knuckles surface shouldn’t be an intentionally displayed, they get uncovered normally. Explicit lines, wrinkles, folds and surface example on the finger knuckle surfaces can be utilized as viable biometric measure individually or in blend with different biometrics. This paper proposes the improvement of a finger knuckle based biometric ID framework. The framework joins Local Binary Pattern (LBP) for feature extraction after preprocessing and upgrading information picture removed from knuckle surface from picture. Furthermore the framework utilizes Bernoulli classifier as a coordinating classifier for individual identification. Neighborhood Binary Pattern is tried, additionally different parameters are removed ...
With the growth of information technology there is a greater need of high security, so biometric ... more With the growth of information technology there is a greater need of high security, so biometric authentication systems are gaining importance. Face recognition is more used because it's easy and non intrusive method during acquisition procedure. Here PCA algorithm is used for the feature extraction. Distance metric or matching criteria is the main tool for retrieving similar images from large image databases for the above category of search. Two distance measures, such as the Manhattan Distance, Mahalanobis Distance have been proposed in the literature for measuring similarity between feature vectors. In content-based image retrieval systems, Manhattan distance and Euclidean distance are typically used to determine similarities between a pair of image. Here facial images of three subjects with different expression and angles are used for classification. Experimental results are compared and the results show that the Mahalanobis distance performs better than the Manhattan Distance.
Using cloud services, anyone can remotely store their data and can have the on-demand high qualit... more Using cloud services, anyone can remotely store their data and can have the on-demand high quality applications and services from a shared pool of computing resources, without the burden of local data storage and maintenance. Cloud is a commonplace for storing data as well as sharing of that data. However, preserving the privacy and maintaining integrity of data during public auditing remains to be an open challenge. In this paper, we introducing a third party auditor (TPA), which will keep track of all the files along with their integrity. The task of TPA is to verify the data, so that the user will be worry-free. Verification of data is done on the aggregate authenticators sent by the user and Cloud Service Provider (CSP). For this, we propose a secure cloud storage system which supports privacy-preserving public auditing and blockless data verification over the cloud.
Ijca Proceedings on Emerging Trends in Computer Science and Information Technology Etcsit1001, Apr 21, 2012
Many Algorithms for implementation of face recognition are popular in face recognition all having... more Many Algorithms for implementation of face recognition are popular in face recognition all having respective advantages and disadvantages. Some improves the efficiency of face recognition, under varying illumination and expression conditions for face images. Feature representation and classification are two key steps for face recognition. Authors have presented novel techniques for face recognition. In this paper, we presented an overview of face recognition techniques and its applications.
IOSR Journal of Computer Engineering, 2014
This study is focused on development an application of an efficient algorithm for subspace cluste... more This study is focused on development an application of an efficient algorithm for subspace clustering and feature extraction that used for classification. Some previous techniques for subspace clustering make use of rank minimization and sparse based such as SSC and LRR are computationally expensive and may degrade clustering performance. So the need is to solve the problems of existing techniques, an algorithm is introduced known as Fixed-Rank Representation i.e. FRR based on matrix factorization is used for subspace clustering and feature extraction and that features are used for classification. For classification we use k-nearest neighbour and neural network classifiers and compare their accuracy or performance to classify the feature extracted from FRR.
IOSR Journal of Computer Engineering, 2012
Face recognition is a hot research topic in the fields of pattern recognition and computer vision... more Face recognition is a hot research topic in the fields of pattern recognition and computer vision, which has been found a widely used in many applications, such as verification of credit card, security access control, and human computer interface. As a result, numerous face recognition algorithms have been proposed, and surveys in this area can be found. Although many approaches for face recognition have been proposed in the past, none of them can overcome the main problem of lighting, pose and orientation. For a real time face recognition system, these constraints are to be a major Analysis (PCA) .These methods challenge which has to be addressed. In this proposed system, a methodology is given for improving the robustness of a face recognition system based on two well-known statistical modelling methods to represent a face image: Principal Component extract the discriminates features from the face. Preprocessing of human face image is done using Gabor wavelets which eliminates the variations due to pose, lighting and features to some extent. PCA extract low dimensional and discriminating feature vectors and these feature vectors were used for classification. The classification stage uses nearest neighbour as classifier. This proposed system will use the YALE face data base with 100 frontal images corresponding to10 different subjects of variable illumination and facial expressions.
2009 IEEE International Advance Computing Conference, 2009
This paper proposes a face recognition method using the FERET face database. Facial images of two... more This paper proposes a face recognition method using the FERET face database. Facial images of two classes and three classes with different expressions and angles are used for classification. Fisher Discriminant Method is used for comparison of the results of two classes with the results of three classes. Euclidian distance method is used for similarity measure. The experimental results have been demonstrated that performance of Fisher Discriminant Analysis for three classes is same as the performance for two classes.
IZVESTIYA SFedU. ENGINEERING SCIENCES
Ijca Proceedings on National Conference on Innovative Paradigms in Engineering Technology 2013, Dec 26, 2013
Neural networks provide an obvious technique for classification. In this paper, neural network ap... more Neural networks provide an obvious technique for classification. In this paper, neural network approach is used for classification of Loan applications. Data has been collected from publicly available source, Alyuda Research, Inc.[1].It does not guarantee the accuracy of the data. This data is intended solely for experimental purposes. FFNN and RBN are used for classification. PREMNMX matlab Function Preprocesses data so that minimum is-1 and maximum is 1
Uploads
Papers by shubhangi sapkal